A Step-by-Step Guide to Real-Time Pricing

An advanced AI model considers much more than what competitors are charging.

by

Marshall Fisher,
Santiago Gallino,

and

Jun Li

by

Marshall Fisher,
Santiago Gallino,

and

Jun Li

From the Magazine (November–December 2023) · Long read

Butenkow/Getty Images

Summary.

In today’s fast-paced world of digital retailing, the ability to revise prices swiftly and on a large scale has emerged as a decisive differentiator for companies. Many retailers now track competitors’ prices via systems that scrape rivals’ websites and use this information as an input to set their own prices manually or automatically. A common strategy is to charge X dollars or X percent less than a target competitor. However, retailers that use such simple heuristics miss significant opportunities to fine-tune pricing.

Some companies are now applying machine-learning models to guide their pricing decisions, but even these retailers tend to take an overly limited approach. They try to match or undercut competitors’ prices without taking into account factors such as whether rivals are out of stock or how consumers make their purchasing decisions.

In this article, the authors present a step-by-step process for dynamic pricing that focuses on building computer models that consider not just competitor pricing but also product availability and customer behavior to recommend optimal prices in real time.

Idea in Brief

The Problem

The ability to use AI to change prices frequently online and in physical stores has become critical to competing in retailing. But even retailers who have built such computer models take an overly limited approach.

What Retailers Get Wrong

They try to match or undercut competitors’ prices without taking into account whether rivals are out of stock or other factors that drive consumers’ purchasing decisions.

The Solution

Build and implement computer models that analyze historical sales data, capture crucial patterns, and consider not just competitor pricing but also product availability and customer behavior to recommend optimal prices in real time.

For digital retailers, the ability to revise prices swiftly and on a large scale has emerged as a decisive differentiator—especially during periods of inflation, when prices fluctuate more frequently. Many retailers scrape rivals’ websites for price information and use it to set their own prices manually or automatically, often using a strategy of charging X dollars or X percent less than the lowest-price competitor. However, retailers that use such simple heuristics miss significant opportunities because they fail to tailor their responses to product availability and demand, among other factors.

A version of this article appeared in the November–December 2023 issue of Harvard Business Review.

MF

Marshall Fisher is the UPS Professor and a codirector of the Fishman-Davidson Center for Service and Operations Management at the University of Pennsylvania’s Wharton School.
SG

Santiago Gallino is the Charles W. Evans Distinguished Faculty Scholar and an associate professor in the Operations, Information, and Decisions Department at the Wharton School.
JL

Jun Li is the Michael R. and Mary Kay Hallman Fellow and an associate professor of technology and operations at the University of Michigan’s Ross School of Business.

A Step-by-Step Guide to Real-Time Pricing

Idea in Brief

The Problem

What Retailers Get Wrong

The Solution

Partner Center

Explore HBR

HBR Store

About HBR

Manage My Account

Follow HBR